622 UnpDERWOOD: THE GENUS GYMNOGRAMME 
following almost exactly the lead of the earlier Species Filicum. It 
is not too much to say that for the pteridophytes it represents a 
system as antiquated and artificial as the so-called sexual system 
of Linnaeus for flowering plants. 
As might be expected Synopsis Filicum is moulded on the old 
"conceptual idea that a genus is a definition framed in set characters, 
instead of the modern evolutionary idea that a genus is a group 
of closely related species. This of necessity reduces the author's 
position to the inclusion of notorious absurdities. Having defined 
Polypodium as with “ Sori on the back of the lobes, round or rarely 
oblong, not more than twice as long as broad,” it becomes neces- 
sary to include in one genus such diverse plants as Dipteris, Phegop- 
teris Dryopteris, Arthropteris tenella, Cyclophorus acrostichoides, and 
Cheiropteris palmatopedata in the same genus with Polypodium pul- 
4re irrespective of any other character than the possession of 
round and naked sori! 
Habit also plays little part in the arrangement of genera and 
subgenera in the Hookerian system. Note if you will a striking 
example presented in the section Hemidictyum of the immens® 
and cumbrous genus Asplenium. This is characterized as having 
“« Veins anastomosing towards the margin. Sori single” and 
among its four species two such absolutely diverse plants as Ceter- 
ach* and Hemidictyum marginatum are included. If we should 
search the whole range of the Polypodiaceae we could scarcely find 
two species more remote in their natural affinities, and yet here 
they stand not only in the same genus but in the same subgenvs 
placed almost side by side! There is nothing more astoundingly 
incongruous in the whole Linnaean system. 
Among the many incongruities among genera, the genus Gre 
nogramme is one of the most irrational and unnatural conglomer” 
ations that appear in the Synopsis Filicum. It forms the largest 
genus of the tribe Grammitideae of the Hookerian arrangement 
* The near allies of Ceterach which forms a distinct genus whose name . agi 
ent on the settlement of the question of generic types, are strangely scattered = ee 
opsis Filicum, In addition to the common European species they are panes ee a 
mans which there appears under 3 Euasplenium.- A. aureum which page 
subspecies of 4. Ceterach 3 and Gymnogramme cordata which appears 19 . tab- 
tribe as its indusium is wanting. Thus are natural relationships violated by cies 
lishment of artificial lines and genera which are limited by definition only. 
